Numpy aggregate functions sum to get total of elements with different options

Numpy

numpy.sum(a,axis=None,dtype=None,out=None,keepdims, initial, where)

Return sum of elements across given axis.

`a`	array, elements to get the sum value
`axis`	Int (optional ), or tuple, default is None, will sum all the elements. If axis given then across the axis is returned.
`dtype`	data-type( Optional ), Data Type of returned array or value.
`out`	Optional. If given then output to be stored. Must be of same time as of the output
`keepdims`	Bool ( Optional ), output matches to the input array dimension.
`where`	Optional, Elements to include for calculation of Sum
`initial`	Optional, int, Initial value of sum. This value will be added to our final output

We will use these parameters in our examples.

Sample array

You can use randint() to create an array for our examples. Or can use fixed elements to create the array.

import numpy as np
# my_data=np.random.randint(2,high=7,size=(3,3),dtype='int16') 
my_data=np.array([[6, 3, 2], [2, 6, 2], [6, 2, 3]])
print(my_data)

Output

[[6 3 2]
 [2 6 2]
 [6 2 3]]

Axis

Sum of the elements across the axis. Axis of Numpy array sum

print("sum()      : ", my_data.sum())
print("sum(axis=0):", my_data.sum(axis=0))
print("sum(axis=1):", my_data.sum(axis=1))

Output

sum()      :  32
sum(axis=0): [14 11  7]
sum(axis=1): [11 10 11]

dtype

The data type of the output. By default the output will have the dtype of input array.

print("sum(axis=1,dtype=np.int8) : ", my_data.sum(axis=1,dtype=np.int8))
print("sum(axis=1,dtype=np.int32) : ", my_data.sum(axis=1,dtype=np.int32))
print("sum(axis=1,dtype=np.float64) : ", my_data.sum(axis=1,dtype=np.float64))
print("sum(axis=1,dtype=np.complex128) : ", my_data.sum(axis=1,dtype=np.complex128))

Output

sum(axis=1,dtype=np.int8) :  [11 10 11]
sum(axis=1,dtype=np.int32) :  [11 10 11]
sum(axis=1,dtype=np.float64) :  [11. 10. 11.]
sum(axis=1,dtype=np.complex128) :  [11.+0.j 10.+0.j 11.+0.j])

keepdims

If it is set to True ( keepdims=True ) then it will take the dimension of input array.

print("sum(keepdims=True) : ", my_data.sum(keepdims=True))
print("sum(keepdims=False) : ", my_data.sum(keepdims=False))

Output

sum(keepdims=True) :  [[43]]
sum(keepdims=False) :  43

out

Alternative output array, must be of same shape as expected output. Let us first check with axis.

x = np.zeros(3,dtype=int)
print(my_data.sum(axis=0,out=x))
print(x)

Output

[14 11  7]
[14 11  7]

Without using axis

y = np.array(1)
print(my_data.sum(out=y))
print(y)

Output

32
32

Using where

By using where we can say which elements to use and which elements not to use ( by setting True or False ) .

print(my_data.sum(where=[True, False,True]))

Output

Using axis with where

print(my_data.sum(axis=1,where=[True, False,True]))
print(my_data.sum(axis=0,where=[True, False,True]))

Output

[8 4 9]
[14  0  7]

initial

We can assign initial value to our Sum. This will be added to final value.

print(my_data.sum(initial=10))

Output

Numpy mean() max() min()

Python Tkinter Pandas Python- Tutorials

Subscribe to our YouTube Channel here

Numpy sum()

Sample array

Axis

dtype

keepdims

out

Using where

initial

Subscribe